Using social psychology to motivate contributions to online communities

Citation: Gerard Beenen, Kimberly Ling, Xiaoqing Wang, Klarissa Chang, Dan Frankowski, Paul Resnick, Robert E. Kraut (2004) Using social psychology to motivate contributions to online communities. Proceedings of the 2004 ACM conference on Computer Supported Cooperative Work (RSS)
DOI (original publisher): 10.1145/1031607.1031642
Semantic Scholar (metadata): 10.1145/1031607.1031642
Sci-Hub (fulltext): 10.1145/1031607.1031642
Internet Archive Scholar (search for fulltext): Using social psychology to motivate contributions to online communities
Tagged: Computer Science (RSS) CSCW (RSS)

Summary

Beenen et al. (2004) was published in CSCW and explores both an example of using findings from social psychology to encourage contributions and reflects on some of the limits of social psychology in this perspective. The paper uses empirical data from a series of experiments in MovieLens, an online collaborative movie rating website associated with the Community Lab project. The paper begins with a rich literature review on the problems and importance of soliciting contributions from users of system and frames their exploration in terms of what social psychologists refer to as "social loafing" which describes how people often exert less effort on a collective task than they do on an individual task.

The first study aims to explore how contributions might be increased through emphasizing the unique value of an individual contribution and through the benefit that contributors, other users, or both will derive. Four hypothesis are leveled that emphasizing each of these qualities will increase contributions. To test these, users were randomized and sent emails soliciting contributive with texts that emphasized each of these qualities. Contribution rates were then measured and compared. The findings were surprising. The emphasize on uniqueness of contributions increased contributions. However, the emphasis on either individual or group benefit decreased contribution rates. The emphasis on both types of benefit together was also lower, but higher than either one separately. There was no interaction between the uniqueness and benefit treatments. The authors admit that the results are confusing.

A second study explores the social psychological phenomena of goal setting and tests if setting contributions goals leads to increased contribution. They hypothesize that numeric goals will result in more contributions, that group goals will be less effective than individual goals, and that goal effectiveness will drop when the goals are too hard (i.e., they'll see a curvilinear relationship). Findings are that the first two hypotheses are confirmed and that the non-linear relationship is very tentatively supported.

Possibly, the most interesting takeaway from the work is perhaps in the discussion where the authors reflect on how reasonable it may be to use design systems and real communities to test social psychology. In particular, they cite failures in implementation, mismatches between scientific and engineering goals and disciplines, and incomplete theories as explanations for unusual result. The take-aways are reflections but they are interesting and informative for anyone interesting in testing social science through experiments in real communities.

Theoretical and Practical Relevance

The paper is reasonably highly cited in CSCW. It provides a good example both of the question of encouraging contributions to online communities and on the limits of testing social psychology or applying its findings to real communities.